Riverbed Technology, Inc.

Opnet Technologies, Inc.

Paul Janes

Karthikeyan Chandrashekar

Alain J Cohen

Pradeep Singh

David Boyd

Walt Whimpenny

Majid Raissi-Dehkordi

Stefan Znam

Ibrahim Utku Moral

Embodiments provide systems, methods, and computer program products to generate a network topology. Internet Protocol (IP) addresses may be collected that immediately precede a first IP address in a set of IP-address-sequences to obtain a first set of previous-hop IP addresses, where each IP-address-sequence in the set of IP-address-sequences comprises a sequence of IP addresses traversed by at least one packet. Next, each IP address in the first set of previous-hop IP addresses may be associated with a first logical node in the network topology.

Network topology generation using traceroute data

Embodiments provide systems, methods, and computer program products for inferring node and link information from traceroute data in order to generate topology information. A system receives traceroute data for a data packet that traverses a path from a source to a destination. The system infers port types for the addresses in the traceroute data and groups subsets of the addresses in the traceroute data into logical nodes based on neighbor relationships demonstrated in backward and forward neighbors sets. The system then generates node and link information based on the inferred and grouped information.

In a network that includes static bandwidth and dynamic bandwidth links, traffic flow at the OSI network layer is simulated at a traffic-flow level at interfaces to fixed bandwidth links, and simulated at a discrete-packet level at interfaces to dynamic bandwidth links. The resultant discrete-packet reception events at the receiving interface(s) of the dynamic bandwidth link are processed to determine the effective bandwidth/throughput of the link, as well as the allocation of this bandwidth among the individual flows through the link. The discrete-packet level receptions are used to reconstruct the parameters of the traffic flow at the network layer of the receiving interface, and this determined traffic flow is simulated accordingly at the next link, depending upon whether the next link is a static or dynamic bandwidth link.

Modeling and simulating flow propagation in dynamic bandwidth systems

Channel access delays and reception uncertainty are modeled as protocol-independent generic processes that are optimized for improved simulation performance. The generic process components are designed such that each different protocol can be modeled using an arrangement of these components that is specific to the protocol. In this way, speed and/or accuracy improvements to the generic process components are reflected in each of such protocol models. If an accurate analytic model is not available for the generic process component, a prediction engine, such as a neural network, is preferably used. The prediction engine is trained using the existing detailed models of network devices. Once trained, the prediction engine is used to model the generic process, and the protocol model that includes the generic component is used in lieu of the detailed models, thereby saving substantial processing time.

Growing community of inventors

Bethesda, MD, United States of America

Paul Janes